Contents lists available at ScienceDirect

Aquaculture

journal homepage: www.elsevier.com/locate/aquaculture

Gene expression of Na+/K+-ATPase α-isoforms and FXYD proteins andpotential modulatory mechanisms in euryhaline milkfish kidneys uponhypoosmotic challenges

Wen-Kai Yanga,b,c, Tse-Lih Chaoa,1, Hsin-Ju Chuanga,b,1, Yao-Chung Hua,b,Catherine Lorin-Nebeld, Eva Blondeau-Bidetd, Wen-Yi Wua, Cheng-Hao Tanga,b,e,Shu-Chuan Tsaif,⁎⁎, Tsung-Han Leea,b,⁎

a Department of Life Sciences, National Chung Hsing University, Taichung 40227, Taiwanb The iEGG and Animal Biotechnology Center, National Chung Hsing University, Taichung 40227, Taiwanc Bachelor Degree Program in Animal Healthcare, Hungkuang University, Taichung 43302, TaiwandUniversité de Montpellier, UMR 9190 MARBEC (UM,-CNRS-IRD-IFREMER), cc 092, Place E. Bataillon, 34095 Montpellier, cedex 5, Francee Department of Oceanography, National Sun Yat-Sen University, Kaohsiung 80424, TaiwanfDepartment of Nursing, Central Taiwan University of Science and Technology, Taichung 40601, Taiwan

A R T I C L E I N F O

Keywords:Na+/K+-ATPaseα-isoformFXYD proteinKidneyMilkfishSalinity

A B S T R A C T

Euryhaline milkfish (Chanos chanos) is an important economic species in Taiwan and is able to survive in bothseawater (SW) and fresh water (FW). Upon salinity changes, euryhaline fish kidneys play a crucial role inosmoregulation. During this process, Na+/K+ -ATPase (NKA) mainly functions in providing the driving forcefor the other ion-transporting systems. In the milkfish, NKA activities have been widely studied, yet the isoformsand modulatory mechanisms of the NKA α-subunits are unknown. Hence, this study was undertaken to revealpotential modulatory mechanisms, including the isoform composition of NKA α-subunits and FXYD proteins (themodulators of NKA), of NKA expression/activity in milkfish kidneys in response to salinity changes. Four iso-forms (CcNKA α1a, α1c, α2, and α3) of NKA α-subunits with distinct tissue distribution were identified inmilkfish. Among them, Ccnka α1a and Ccfxyd2 were predominantly expressed in the kidney. Upon salinitychallenge, parallel expression patterns were found in renal Ccnka α1a and Ccfxyd2, showing an increase in theFW-acclimated milkfish but not in the SW-acclimated group. After transfer, no matter from SW to FW or FW toSW, most of the renal Ccnka and Ccfxyd genes (particularly Ccnka α1a and Ccfxyd2) significantly changed.Overall, these results suggest that the NKA activity of milkfish kidneys may be modulated via CcNKA α1a andCcFXYD2 for osmoregulation in response to salinity challenges.

1. Introduction

Milkfish (Chanos chanos), a marine teleost that resides in tropicalregions from the Pacific Ocean to the Indian Ocean, is a commerciallyimportant aquaculture species in Southeast Asian countries, includingTaiwan (Martinez et al., 2006). Milkfish can survive in hypoosmotic(i.e., fresh water [FW]) or hyperosmotic (e.g., seawater [SW]) en-vironments (Bagarinao, 1994; Martinez et al., 2006). In Taiwan, milk-fish are cultured in ponds with water of various salinities including FW,

brackish water, or SW (Martinez et al., 2006). Hence, for elucidatingthe effects of environmental factors (such as salinity) on physiologicalmechanisms as the background knowledge to improve aquaculturepractice of milkfish farm, further investigation of their molecular me-chanisms of osmoregulation in response to ambient salinity changes iswarranted (Lin et al., 2003). There is evidence that the milkfish is ableto maintain internal homeostasis, such as muscle water content, os-molality of the plasma and urine, and concentrations of plasma Na+

and Cl−, within narrow physiological ranges under different

https://doi.org/10.1016/j.aquaculture.2019.01.046Received 8 November 2018; Received in revised form 18 January 2019; Accepted 19 January 2019

Abbreviations: Cc, Chanos chanos; FW, freshwater/fresh water; NKA, Na+/K+-ATPase; PCR, polymerase chain reaction; PK, protein kinase; QPCR, quantitative real-time PCR; SE, standard error; SW, seawater; TM, transmembrane domain.

⁎ Correspondence to: T. H. Lee, Department of Life Sciences, National Chung Hsing University, 145 Xingda Road, Taichung 40227, Taiwan.⁎⁎ Correspondence to: S. C. Tsai, Department of Nursing, Central Taiwan University of Science and Technology, 666 Buzih Road, Taichung 40601, Taiwan.E-mail addresses: sctsai@ctust.edu.tw (S.-C. Tsai), thlee@email.nchu.edu.tw (T.-H. Lee).

1 T.L. Chao and H.J. Chuang were equally contributed to this paper.

Aquaculture 504 (2019) 59–69

Available online 23 January 20190044-8486/ © 2019 Elsevier B.V. All rights reserved.

T

environmental salinity conditions (Kang et al., 2015; Lin et al., 2003;Tang et al., 2010). Thus, milkfish is a good experimental species forstudies on osmoregulation.

Teleosts maintain relatively stable internal environments by theactivation of osmoregulatory organs, including the gill, kidney, and gut(Marshall and Grosell, 2006; Takei et al., 2014; Whittamore, 2012).Among these organs, the kidney plays an important role in osmor-egulation in both SW and FW teleosts, performing completely differentfunctions in these two environments (Madsen et al., 2015; Marshall andGrosell, 2006; Takei et al., 2014). In SW residents, the hypertonic en-vironment results in passive loss of water and influx of excess ions. Theythus drink SW to prevent dehydration and actively excrete ions throughthe kidneys and gills. In contrast, FW teleosts gain excess water loadingand lose ions to their surrounding hypotonic environment. For com-pensation, the primary function of fish kidneys is to excrete excesswater and reabsorb filtered solutes (Madsen et al., 2015; Marshall andGrosell, 2006). In these processes, renal Na+/K+-ATPase (NKA) plays acrucial role in ionoregulation/osmoregulation for cellular homeostasisof ions and water (Marshall and Grosell, 2006; Whittamore, 2012).

NKA is a primary driving force for other ion-transporting systems inthe osmoregulatory organs, including the kidneys (Hwang et al., 2011;Whittamore, 2012). It is a P-type ATPase and consists of catalytic α-subunits and smaller glycosylated β-subunits (Morth et al. 2011;Toyoshima et al. 2011). Among them, the catalytic α-subunit, con-taining the binding sites of Na+, K+, and ATP, plays a crucial role inNKA function, including osmoregulation (Evans and Claiborne, 2009;Tang et al., 2009). In birds and mammals, four α-isoforms (α1, α2, α3,and α4) of NKA with tissue-specific distribution have been reported(Blanco, 2005; Morth et al., 2011; Suhail, 2010). Among these isoforms,α1 is expressed ubiquitously in transporting epithelia for maintainingosmotic balance and cell volume regulation. Meanwhile, α2 is mainlyfound in the brain, muscle, and heart; α3 is present primarily in theneurons and ovaries; and α4 is expressed only in the testis. The switchof these α-isoforms is considered to meet specific requirements fordifferentiated cell-specific functions of cation transport (Blanco, 2005;Blanco and Mercer, 1998; Suhail, 2010). In teleosts, only three α-iso-forms (α1, α2, and α3) but numerous paralogs have been identified(Armesto et al., 2014; Bollinger et al., 2016; Ching et al., 2015;Richards et al., 2003; Wong et al., 2016). As in mammals, tissue-specificexpression of these teleostean α-subunits has been demonstrated.Moreover, in the osmoregulatory organs, the expression (mRNA and/orprotein) of the predominant NKA α-isoforms (especially α1 and itsparalogs) is salinity-dependent, indicating their roles in fish osmor-egulation (Armesto et al., 2014; Bollinger et al., 2016; Richards et al.,2003; Wong et al., 2016).

FXYD proteins identified in some teleosts are considered to be im-portant modulatory mechanisms for NKA activity (Saito et al., 2010;Tipsmark, 2008; Yang et al., 2013). At present, at least 12 members ofthe FXYD protein family (i.e., FXYD1-12) have been reported inmammals and fish, and most FXYD proteins exhibit distinct tissue dis-tribution (Geering, 2008; Hwang et al., 2011; Yang et al., 2013). TheFXYD protein family has been more widely studied in mammals andelasmobranchs compared to teleosts (Crambert and Geering, 2003;Geering, 2008). Recently, salinity effects on expression of FXYD pro-teins were investigated in teleostean osmoregulatory ograns (Tipsmark,2008; Wang et al., 2017; Yang et al., 2018). The potential function ofsome FXYD proteins upon osmotic challenge has been demonstrated ina limited number of teleostean species (Chang et al., 2016; Saito et al.,2010; Wang et al., 2017; Yang et al., 2016b). To date, however, theroles of most FXYD proteins are not yet clear in teleosts.

In the milkfish, the effects of salinity on NKA expression, includingNKA activity (Kang et al., 2015; Lin et al., 2003; Tang et al., 2010) andpotential isoform composition of NKA α-subunits (protein levels only)(Tang et al., 2009; Yang et al., 2016a), have been investigated. Mole-cular characteristics and modulatory mechanisms of NKA α-subunitisoforms, however, are not clear. Hence, this study focused on potential

modulatory mechanisms, including isoform composition of NKA α-subunits and FXYD proteins, on NKA expression/activity in response tosalinity changes. The goals of the present study were to (1) identify theisoforms of milkfish NKA α-subunits; and (2) investigate the gene ex-pression of NKA α-subunits and FXYD protein in milkfish kidneys uponhypoosmotic challenge. This paper sheds a light on the molecularmodulatory mechanisms of NKA expression of euryhaline teleosts fol-lowing salinity challenge.

2. Materials and methods

2.1. Ethical approval

The protocols employed for the experimental fish were reviewedand approved by the Institutional Animal Care and Use Committee ofthe National Chung Hsing University (IACUC approval no. 105-130),and were carried out in accordance with the approved guidelines.

2.2. Experimental animals and environments

Juvenile milkfish (C. chanos) with a standard length of 9–11 cmwere purchased from the Da-Shun milkfish aquafarm in Tainan,Taiwan. Fish were transported to the laboratory, where they were ac-climated to aerated dechlorinated local tap water (i.e., FW) or SW(35‰) at 28 ± 1 °C under a 12-h light: 12-h dark cycle. SW was pre-pared from FW by adding standardized amounts of synthetic sea salt(Blue Treasure Synthetic Sea Salt; Blue Treasure, Surry Hills, Australia).The details of water parameters were reported in our previous study(Yang et al., 2017). Water was continuously circulated through fabric-floss filters. Fish were fed twice a day with a diet of commercial pellets.

2.3. Acclimation and transfer experiments

For long-term acclimation, the milkfish were acclimated to SW orFW for at least 4 weeks. In the time-course experiments, after long-termacclimation, milkfish in the study groups were transferred directly fromSW to FW, and vice versa. Moreover, fish in the control group weretransferred from SW to SW or from FW to FW for their respective ex-periments. Milkfish were sampled at 1/2, 1, 2, 4, and 7 days post-transfer, based on previous studies (Hu et al., 2017; Tang et al., 2009).Before sampling, the fish were not fed for one day and were then an-esthetized using 39.8 μM phenoxyethanol. The tissues and organs weredissected and directly placed into liquid nitrogen and then stored im-mediately at −80 °C for subsequent analysis.

2.4. Total RNA extraction and reverse transcription

Total RNA from various milkfish tissues/organs was extracted withIsoI-RNA Lysis Reagent (2302700; 5 Prime, Gaithersburg, MD, USA) inaccordance with the manufacturer's instructions. The integrity andconcentration of extracted RNA was verified and measured by 1%agarose gel electrophoresis and a NanoDrop 2000 (Thermo,Wilmington, DE, USA), respectively. For reverse transcription, cDNAswere synthesized with the iScript™ cDNA Synthesis Kit (170–8891; Bio-Rad, Hercules, CA, USA) or SMART™ RACE cDNA Amplification Kit(634858; Clontech, Palo Alto, CA, USA) according to the manufacturer'sinstructions. The RNA and cDNA were stored at −80 °C and− 20 °C,respectively.

2.5. Sequencing and identification of milkfish NKA α-subunits

Genes coding for NKA α-subunits were derived from the milkfishtranscriptome database (Hu et al., 2017). They were confirmed throughRT- (reverse transcription) or RACE- (rapid amplification of cDNA ends)PCR (polymerase chain reaction) and then sequenced by the Tri-Biotech(Taipei, Taiwan).

W.-K. Yang et al. Aquaculture 504 (2019) 59–69

60

To characterize NKA proteins, the amino acid sequences were de-duced from sequenced genes using the NCBI Open Reading FrameFinder (https://www.ncbi.nlm.nih.gov/orffinder/). Subsequently, thetransmembrane domains were predicted using the MEMSAT-SVM pro-vided by the PSIPRED protein structure prediction server (http://bioinf.cs.ucl.ac.uk/psipred/) (Buchan et al., 2013), and comparisons weremade with transmembrane domains of teleostean NKA including theSenegalese sole (Solea senegalensis; Armesto et al., 2014) and Europeansea bass (Dicentrarchus labrax; Blondeau-Bidet et al., 2016). Potentialphosphorylation sites of protein kinase C (PKC) and A (PKA) werepredicted via NetPhos 3.1 (http://www.cbs.dtu.dk/services/NetPhos/)(Blom et al., 2004). The binding sites of Na+ and K+ were predictedaccording to the homology modeling of the cation binding sites ofhuman NKA (Ogawa and Toyoshima, 2002).

For identification of these protein sequences, multiple sequencealignments, phylogenetic analysis, and protein identity were per-formed. Amino acid sequences for NKA α-subunits from different or-ganisms (Tables S1 and S2) were aligned by MUSCLE (multiple se-quence comparison by log-expectation) using MEGA X (Kumar et al.,2018) for the subsequent analyses. These aligned sequences were usedto confirm the positions of α1a-like and α1b-like residues based onBlondeau-Bidet et al. (2016) in order to distinguish the α1 paralogs bycomparison with sequences of other teleostean species (Fig. S1). Aphylogenetic tree was constructed by the maximum likelihood methodwith 1000 bootstraps using MEGA X. Moreover, amino acid identityamong teleosts and mammals was assessed using BioEdit 7.2.5 with theSequence Identity Matrix (Hall, 1999).

2.6. Quantitative real-time PCR (QPCR)

The methods, including PCR protocol and data calculation, weredescribed previously (Chang et al., 2016; Hu et al., 2017). Expressionlevels of target genes were quantified using 2× FastStart UniversalSYBR Green Master (04913850001; Roche, Mannheim, Germany) withthe MiniOpticon real-time PCR system (Bio-Rad). Then, mRNA valuesfor the target genes were normalized using the expression of β-actinmRNA from the same cDNA sample (Tang et al., 2010). Primer in-formation used is listed in Table S2. The amplification efficiency wassimilar for all primer pairs used in this study (range: 0.93–1.08). Toconfirm the specificity of these amplifications from each test sample,sequencing, melting curve analysis, and electrophoresis were carriedout. For each unknown sample, relative fxyd gene expression was ob-tained using the formula 2^−[(Ct target genes, n− Ct β-actin, n)− (Ct target

genes, c− Ct β-actin, c)], where Ct corresponds to the threshold cyclenumber (Livak and Schmittgen, 2001). The letter “n” indicates eachcDNA sample used in these experiments and “c” indicates the controlmixed with cDNAs from many organs.

2.7. Statistical analysis

All values are shown as the mean± standard error (SE). TheShapiro-Wilk normality test was applied. Because some of the data werenot normally distributed, non-parametric tests were used. Differences inthe mRNA levels between groups were analyzed via the Mann-WhitneyU test or Kruskal-Wallis test for pair-wise or multiple comparisons, re-spectively. All analyses were performed using SPSS 20 software (SPSS,Chicago, IL, USA), and P values of< 0.05 were considered to be sig-nificant.

3. Results

3.1. Characterization of milkfish NKA α-isoforms

Four NKA α-isoforms, CcNKA α1a, α1c, α2, and α3, were identifiedin milkfish (Fig. 1; Table S2). Among them, CcNKA α2 was a partialsequence because only 952 deduced amino acids were obtained. These

amino acid sequences of four CcNKA α-isoforms shared similar char-acteristic features, including transmembrane domains (TMs), phos-phorylation sites of PKC and PKA binding sites, binding sites for Na+

and K+, and a conserved C-terminus (Fig. 1). For further confirmation,a multiple sequence alignment of teleostean NKA α1-isoforms wascarried out (Fig. S1). The results revealed that CcNKA α1a belongs toone of the teleostean NKA α1a groups, because it has more α1a-likeresidues (blue triangles) as opposed to α1b-like residues (green trian-gles). On the other hand, in the phylogenetic analysis, amino acid se-quences of NKA α-subunit paralogs from teleosts, chicken, and human,with the sea urchin as outgroup, were aligned and a phylogenetic treewas constructed (Fig. 2). These proteins were divided into four groups(α1a/b, α1c, α2, and α3), and all CcNKA α-isoforms of milkfish wereclassified into their respective groups. Among these clades, moreover,close relationships were noted among the α1a/b and α1c of teleosts andthe α1 of human and chicken. Additionally, higher identities of bothcDNA and amino acid sequences were found between CcNKA α1a orα1c and α3 rather than α2 (Table 1). Moreover, the sequence identitiesamong all CcNKA α-isoforms were higher for protein than for cDNA(Table 1). The comparisons of amino acid sequences of CcNKA α1a, α1c(Table 2), α2, and α3 (Table 3) individually with the homologues fromother animal species showed that they shared a higher degree ofhomology with teleosts than with bird and mammal. Among the se-quences of NKA α-isoforms, CcNKA α2 revealed a lower identity withother teleostean NKA α2 (83.5–85.6%; Table 3).

3.2. Tissue distribution of milkfish Ccnka α genes

The QPCR results revealed tissue-specific expression for most of theCcnka α genes (Fig. 3). Ccnka α1a was present predominantly in theintestine and kidney, meanwhile Ccnka α1c was mainly expressed in thebrain, heart, and liver. Ccnka α2 was expressed pervasively in all de-tected organs, while Ccnka α3 was primarily found in the brain andliver.

3.3. Salinity effects on Ccnka α and Ccfxyd genes in kidneys of SW and FWmilkfish

In milkfish kidneys, Ccnka α1a was most predominantly expressedamong Ccnka α genes at both salinities, followed by Ccnka α1c(Fig. 4A). Meanwhile, renal Ccfxyd2 mRNA abundance was the highestin both SW and FW groups, followed by Ccfxyd8 and Ccfxyd9 (Fig. 4B).Low mRNA levels of Ccnka α2, Ccnka α3, Ccfxyd7a, Ccfxyd7b, andCcfxyd11 were observed in milkfish kidneys. Moreover, the expressionof Ccnka α1a, Ccfxyd2, Ccfxyd7b, and Ccfxyd9 genes in milkfish kidneyswas salinity-dependent (Fig. 4). Renal mRNA levels of Ccnka α1a,Ccfxyd2, and Ccfxyd9 were significantly higher in the FW group than inthe SW group. In contrast, a higher mRNA abundance of renal Ccfxyd7bwas found in the SW-acclimated milkfish.

3.4. Time-course changes in the Ccnka α and Ccfxyd mRNA levels ofmilkfish kidneys upon salinity challenge

In milkfish kidneys, after transfer from SW to FW, the mRNA pro-files of Ccnka α and Ccfxyd genes were distinct, divided roughly intothree patterns (Fig. 5). (1) mRNA expression of Ccnka α1a, Ccfxyd2, andCcfxyd8 first changed (increased/decreased) and then recovered, com-pared to either the initial time-point (day 0) or the control group at thesame time-point (Fig. 5A, C, D). (2) mRNA levels of Ccfxyd9 quicklyincreased and then slightly decreased. Expression of Ccfxyd9mRNA wasnot significantly different from the initial time-point but was sig-nificantly higher than that of the control group at 7 days post-transfer(Fig. 5E). (3) No difference was observed in Ccnka α1c mRNA expres-sion among different time-points or between the study and controlgroup of each time-point after transfer (Fig. 5B). Among these profiles,most of the peak values were observed within 2 days post-transfer.

W.-K. Yang et al. Aquaculture 504 (2019) 59–69

61

Fig. 1. Alignment of putative amino acid sequences of milkfish NKA (CcNKA) α-isoforms. Dashes represent gaps, and the predicted transmembrane domains (TMs)are shown in green underlines. Vertical boxes represent coordinating residues for Na+ (Na) or K+ (K) binding and for either Na+ or K+ binding (Na/K). Potentialphosphorylation sites are shown with crosses (+), asterisks (*), and hash tags (#) for different kinases, including protein kinase C, protein kinase A, and both,respectively. Triangles indicate the KETYY motif. ECD, extracellular domain; ICD, intracellular domain. (For interpretation of the references to colour in this figurelegend, the reader is referred to the web version of this article.)

W.-K. Yang et al. Aquaculture 504 (2019) 59–69

62

On the other hand, after transfer from FW to SW, mRNA levels ofrenal Ccnka α1a were clearly downregulated (Fig. 6A), while mRNAexpression of renal Ccnka α1c and Ccfxyd2 was first changed (in-creased/decreased) and then recovered (Fig. 6B, C). In renal Ccfxyd8

mRNA abundance, a significant change was found between 1 and 2 dayspost-transfer, and there was no difference among the other time-points(Fig. 6D). Among these profiles, mRNA levels of renal Ccnka α1a andCcfxyd2 clearly decreased at 1 day post-transfer, and the peak value ofCcnka α1c mRNA was observed at 4 days post-transfer. Moreover, nodifference was observed in Ccfxyd9 mRNA expression of milkfish kid-neys among time-points or between the study and control group of eachtime-point after transfer (Fig. 6E).

Meanwhile, in both control groups (transfer from SW to SW andfrom FW to FW), no difference was found among time-points aftertransfer.

Fig. 2. Phylogenetic analysis of putative amino acid sequences of NKA α-isoforms using the maximum likelihood method. The numbers at the forks indicate thebootstrap values (%). Sea urchin NKA α was used as the outgroup for the root tree. Accession numbers are listed in Tables S1 and S2.

Table 1Percentage of cDNA (open reading frame only; above diagonal) and amino acid(below diagonal) sequence identity among milkfish Na+/K+-ATPase (CcNKA)α-subunits.

CcNKA α1a(%)

CcNKA α1c(%)

CcNKA α2 (%) CcNKA α3 (%)

CcNKA α1a – 82.0 72.4 75.5CcNKA α1c 91.4 – 70.6 78.4CcNKA α2 79.9 81.1 – 72.8CcNKA α3 85.4 86.0 80.8 –

Accession numbers are listed in Table S2.

W.-K. Yang et al. Aquaculture 504 (2019) 59–69

63

4. Discussion

4.1. Milkfish NKA α-isoforms

To further verify the modulatory mechanisms and molecular re-sponses of milkfish NKA in response to salinity changes, we identifiedthe cDNA sequences of nka α-subunits of milkfish for the first time. Fourisoforms of the NKA α-subunits in milkfish, i.e., CcNKA α1a, α1c, α2,and α3, were identified. Among them, the amino acid sequences ofCcNKA α1a and α1c were the most similar compared to α2 and α3.Similar results (highly similar α1 paralogs) were found in other teleosts(Armesto et al., 2014; Richards et al., 2003). A previous study indicatedthat fish α1 paralogs are derived from an ancestral α1. The ancestral α1was first duplicated into α1a/b and α1c, and then α1a/b was dupli-cated into α1a and α1b (Wong et al., 2016). The grouping of α1a andα1b in the phylogenetic trees of previous studies was sometimes con-fusing because of high similarities between their sequences (Bollingeret al., 2016; Ip et al., 2012; Wong et al., 2016). In this study, only onegene of the α1a/b paralogs was cloned from milkfish. However, itcannot be clearly identified from α1a or α1b via phylogenetic analysis

and protein identity as compared with other homologues. Therefore,according to Blondeau-Bidet et al. (2016), a multiple sequence align-ment was performed. Subsequently, the gene was named Ccnka α1abecause of its higher similarity of amino acid sequences with otherteleostean NKA α1a. In addition to CcNKA α1a, other CcNKA α-sub-units could be named via the results of phylogenetic analysis and pro-tein identity, referring to the nomenclature from previous studies(Armesto et al., 2014; Blondeau-Bidet et al., 2016).

Previous studies indicated that amino acid sequences of NKA α-subunits are highly conserved across animal species (Armesto et al.,2014; Martin, 2005). In this study, higher similarities (over 80 and 90%for α2 and others, respectively) of amino acid sequences among CcNKAα-isoforms and their respective homologues from other animal specieswere obtained. Likewise, similar results were found in other fish, in-cluding the Senegalese sole (Armesto et al., 2014), electric eel (Elec-trophorus electricus; Ching et al., 2015), and European seabass(Blondeau-Bidet et al., 2016). However, in the climbing perch (Anabastestudineus), NKA α1a showed 70–80% homology to most NKA α1-isoforms from other teleosts (Ip et al., 2012). These results revealed thatthere might still be certain differences among NKA α homologues. Onthe other hand, in a protein structure analysis, four CcNKA α-isoformsshowed the typical topology of P-type ATPase, including 10 TMs withion-binding sites, five extracellular domains, and four intracellulardomains containing phosphorylation sites of PKC and PKA, as seen inmammalian NKA α-subunits (Blanco, 2005; Martin, 2005). Thesecharacteristic features have also been found in other teleosts (Armestoet al., 2014; Blondeau-Bidet et al., 2016; Ching et al., 2015). Moreover,the lysine (K)-rich sequences in the N-terminus of CcNKA α-isoforms(except CcNKA α2) present in many teleostean NKA α-subunits areconsidered to play a key role in forming a cation-selective gate via anexcess of positively charged amino acids during certain stages of thetransport process (Blanco and Mercer, 1998; Daly et al., 1994; Vasiletsand Schwarz, 1993) and/or in interacting with other proteins(Toyoshima et al., 2011). The KETYY motif in the tails of CcNKA α-isoforms is well-conserved except for lysine (K), indicating that it iscrucial for Na+ binding (Morth et al., 2011). The conserved degrees ofKETYY motif differ among teleostean NKA α-subunits, and these dif-ferences might cause a difference in Na+ affinity (Blondeau-Bidet et al.,2016; Ching et al., 2015). Taken together, these conserved motifs anddomains of the CcNKA α-isoforms play an important role in membranelocalization and in ionoregulation via phosphorylation and depho-sphorylation, as do other NKA α-subunits.

In NKA α-subunits, the form and number of isoforms vary amongteleostean species. Some teleosts, including zebrafish (Danio rerio; Liaoet al., 2009), rainbow trout (Oncorhynchus mykiss; Richards et al.,2003), Senegalese sole (Armesto et al., 2014), Japanese medaka (Or-yzias latipes; Bollinger et al., 2016), and Japanese eel (Anguilla japonica;Wong et al., 2016), have more NKA α-subunit isoforms than milkfish. Inteleosts, moreover, most of these subunits belong to the highly di-versified α1-isoforms, with tissue-specific distribution (Armesto et al.,2014; Wong et al., 2016). These two findings were also observed inmilkfish, which showed two α1-isoforms and distinct tissue distributionpatterns of four Ccnka α genes. As in other teleosts, Ccnka α1a is presentprimarily in the osmoregulatory organs, while Ccnka α1c is expressedubiquitously in milkfish. However, their distribution patterns and ex-pression levels vary by species (Armesto et al., 2014; Blondeau-Bidetet al., 2016; Ip et al., 2012; Richards et al., 2003; Wong et al., 2016). Inthe Senegalese sole, the nka α1a is mainly expressed in the gill, kidney,and intestine (Armesto et al., 2014), while in the rainbow trout nka α1ais present primarily in the gill (Richards et al., 2003). On the otherhand, Ccnka α2 mRNA are expressed ubiquitously in all detected or-gans, yet their abundance is unexpectedly low as compared to that ofother Ccnka α genes. In mammals and other teleosts, nka α2 is expressedpredominantly in the muscle (Armesto et al., 2014; Bollinger et al.,2016; Richards et al., 2003; Suhail, 2010). Similar to results found inother animal species, Ccnka α2 mRNA was present primarily in milkfish

Table 2Amino acid identity (%) among NKA α1-isoforms.

Classification Species Milkfish

α1a (%) α1c (%)

Teleost Killifish α1a 92.2 90.7European seabass α1a 91.5 90.4European seabass α1b 89.2 92.5Senegalese sole α1a 91.9 90.5Senegalese sole α1b 89.6 93.1Stickleback α1a 89.1 88.2Stickleback α1b 88.6 91.8Japanese medaka α1a 88.5 87.5Japanese medaka α1b 90.1 88.9Japanese medaka α1c 88.8 92.3Rainbow trout α1a 82.7 82.9Rainbow trout α1b 90.5 88.7Rainbow trout α1c 87.7 91.8Climbing perch α1a 81.2 80.2Climbing perch α1b 86.0 85.9Climbing perch α1c 90.5 89.8Electric eel α1c1 91.0 96.0

Bird Chicken α1 87.5 88.6Mammal Human α1 88.5 89.4

Accession numbers are listed in Tables S1 and S2.

Table 3Amino acid identity (%) among NKA α2- and α3-isoforms.

Classification Species Milkfish

α2 (%) α3 (%)

Teleost Killifish α2 85.6 81.9Killifish α3 81.1 94.1Senegalese sole α2 85.1 81.4Senegalese sole α3a 81.1 93.7Senegalese sole α3b 80.9 90.6Japanese medaka α2 83.5 81.1Japanese medaka α3a 80.9 93.2Japanese medaka α3b 80.2 91.3Rainbow trout α2 84.0 82.0Rainbow trout α3 79.7 89.0Electric eel α2 83.8 80.4Electric eel α3 81.2 94.8

Bird Chicken α2 83.8 84.2Chicken α3 82.8 89.5

Mammal Human α2 83.1 84.3Human α3 82.9 88.9

Accession numbers are listed in Tables S1 and S2.

W.-K. Yang et al. Aquaculture 504 (2019) 59–69

64

muscle in our preliminary study (data not shown). In particular, thetissue distribution patterns of Ccnka α3, which was found mainly in thebrain and liver, were different from those of other teleosts. Some tele-ostean nka α3 are expressed predominantly in the brain alone, as in theSenegalese sole (both α3a and α3b; Armesto et al., 2014). Other tele-ostean nka α3 are ubiquitously expressed in various organs, as in theMozambique tilapia (Oreochromis mossambicus; α3; Feng et al., 2002)and rainbow trout (α3; Richards et al., 2003). Meanwhile, in the swampeel (Monopterus albus), α3a is found only in the brain and α3b is widelydistributed in various organs (Chen et al., 2013). Overall, these distincttissue distribution patterns of four Ccnka α genes might reveal that theyplay different roles in diverse organs as other NKA α-subunits inmammals and other teleosts.

4.2. Salinity-dependent expression of renal Ccnka α and Ccfxyd genes inmilkfish

Both switching of NKA α-subunits (the catalytic subunit) and FXYDproteins are modulatory mechanisms for NKA activity (Geering, 2008;Toyoshima et al., 2011). Recently, the relative abundance of multipleNKA α-isoforms and FXYD proteins was found to be altered with am-bient salinity changes in several species of teleosts (Hwang et al., 2011;Wong et al., 2016; Yang et al., 2018). Most of these studies, however,focused on the gills. Moreover, no study investigated the potentialmodulatory mechanisms for NKA expression/activity by analyzing

isoform compositions of NKA α-subunits and FXYD proteins simulta-neously in teleostean kidneys. In milkfish kidneys, Ccnka α1a andCcfxyd2 might play major roles, since they were highly expressed.Conversely, the low levels of renal Ccnka α2, α3, Ccfxyd7a, Ccfxyd7b,and Ccfxyd11 implied that they might be functional in other organs, likethose of mammals and other teleosts (Blanco, 2005; Geering, 2008;Hwang et al., 2011; Yang et al., 2013). In mammalian NKA α-subunits,NKA α-isoforms possessed different affinities for substrate ions, oua-bain, and ATP (Blanco, 2005). Furthermore, in salmonid gills, NKA α1aand α1b were considered to be the FW- and SW-type isoforms, re-spectively (Dalziel et al., 2014; McCormick et al., 2009; Richards et al.,2003). In the kidneys of SW or FW milkfish, the mRNA abundance ofCcnka α1a was higher than that of other Ccnka α genes, indicating thatthere might be no switching of the NKA α-isoforms, unlike the results infish gills.

Upon salinity challenge, in this study, parallel expression patternswere found in renal Ccnka α1a and Ccfxyd2, showing an increase in FW-acclimated milkfish. Their expression profiles were similar to thosefound in previous studies focused on NKA expression of milkfish kid-neys, including mRNA (pan-α1), protein (both pan-α- and pan-β-sub-units), and enzyme activity (Tang et al., 2010; Yang et al., 2016a). Aprevious study also reported that in milkfish kidneys, protein abun-dance of the NKA α1-isoform was increased in the FW-acclimatedgroup, while no difference was found in NKA α2- and α3-isoforms be-tween the SW and FW groups (Yang et al., 2016a), which is in

Fig. 3. Tissue distribution of four milkfish NKA (Ccnka) α genes. The values are means± SE (N = 8 or 6 for gill/kidney or other organs; combined with equal samplesize from all salinity groups). Different letters indicate significant differences among Ccnka α genes (Kruskal-Wallis test; P < .05). Int., intestine.

Fig. 4. Comparisons of renal Ccnka α mRNA abundance in themilkfish. The values are means± SE (N= 4). Different lettersindicate significant differences among Ccnka α genes in sea-water (SW; letter: a, b) or freshwater groups (FW; letter: x, y)(Kruskal-Wallis test; P < .05). The asterisks (*) indicate sig-nificant differences between the SW and FW groups for thesame gene (Mann-Whitney U test; P < .05).

W.-K. Yang et al. Aquaculture 504 (2019) 59–69

65

accordance with our data. Moreover, there was no difference in Ccnkaα1c mRNA levels between FW and SW. Therefore, in FW-acclimatedmilkfish, induction of renal Ccnka α1a transcripts rather than Ccnka α1cmight contribute to increased protein abundance of the NKA α1-iso-form and then enhance NKA activity. On the other hand, a few studieshave investigated the renal FXYD2 of teleosts, including the Atlanticsalmon (Salmo salar; Tipsmark, 2008) and zebrafish (Saito et al., 2010).In kidneys of Atlantic salmon, fxyd2 mRNA levels were higher in FWfish than in the SW group (Tipsmark, 2008). Although some studieshave focused on other FXYD proteins of teleostean kidneys in responseto salinity changes (Hu et al., 2014; Liang et al., 2017; Tang et al., 2012;Wang et al., 2017; Yang et al., 2016b), the role of renal FXYD2 insalinity challenge is not yet clear. Mammalian FXYD2 is also expressedin the kidney and inhibits NKA activity via binding with NKA(Arystarkhova, 2016; Geering, 2008). In the present study, however, wehave no evidence regarding the potential inhibitory role of CcFXYD2 onNKA activity. In summary, when milkfish were acclimated to FW, highlevels of renal Ccnka α1a and Ccfxyd2 might cooperate to maintain abalance of high NKA activity for ionoregulation. More studies should be

performed in the future to clarify their roles in osmoregulation.To further investigate the potential modulatory mechanisms, time-

course experiments were performed. The levels of renal Ccnka α1a andCcfxyd2, the predominant participators, changed quickly within 1 dayafter both transfer treatments. Similar results (quick changes) were alsofound in the gill NKA expression (both mRNA and protein of pan-α1) ofmilkfish since 1 day after transfer from SW to FW (Hu et al., 2017).Unlike the gill, changes in renal Ccnka α1a and Ccfxyd2 expression wererecovered subsequently. Similar patterns (recovered or unchanged ex-pression) were found in renal Ccnka α1c, Ccfxyd8, and Ccfxyd9 aftertransfer. Furthermore, in milkfish kidneys, at 7 days post-transfer, sig-nificant changes in mRNA abundance were only observed in Ccnka α1a(from FW to SW) and Ccfxyd9 (from SW to FW). These findings of re-covered and stable expression in nka α genes have also been reported inkidneys of other teleosts (Blondeau-Bidet et al., 2016; Wong et al.,2016; Zhu et al., 2018). In kidneys of the European seabass, mRNAlevels of both nka α1a and α1b increased quickly and then recoveredafter transfer from SW to FW (Blondeau-Bidet et al., 2016). However,no difference in expression was found in most renal nka α genes of the

Fig. 5. Time-course changes in mRNA expression of renal Ccnka α genes in the milkfish after transfer from seawater (SW) to fresh water (FW) (the study group) or toSW (the control group). The values are means± SE (N = 4 or 6 in the initial time-point or others, respectively). Different letters indicate significant differencesamong time- points in the study group (from SW to FW; Kruskal-Wallis test; P < .05). No difference was found among time-points in the control group (from SW toSW). The asterisks (*) indicate significant differences between the study and control groups at the same time-point post transfer (Mann-Whitney U test; P < .05).

W.-K. Yang et al. Aquaculture 504 (2019) 59–69

66

Japanese eel after transfer from FW to SW (Wong et al., 2016). Previousstudies reported that renal nka α expression might be modulated in ashort time (Blondeau-Bidet et al., 2016) and that expression patterns oftemporal changes might vary with species (Wong et al., 2016; Zhuet al., 2018). Moreover, the osmoregulatory capabilities of teleosteankidneys might be correlated with their evolutionary histories or naturalhabitats (Yang et al., 2016a,b). In the milkfish, which has excellentosmoregulatory capacity (Tang et al., 2010; Yang et al., 2016a), ourresults showed that genes expressed at the kidney level can be effec-tively regulated. The findings also revealed that the regulatory processmight be divided into two periods, a short-term rapid modulationperiod (within 7 days) and a subsequent moderate regulation period, inresponse to salinity challenge. Other modulatory mechanisms of NKAα-subunits might be involved during this period, such as endocrineregulation and protein phosphorylation (Crambert and Geering, 2003;Toyoshima et al., 2011). Notably, when the SW-residing milkfish weretransferred to a hypoosmotic environment (FW), changes in renal Ccnkaand Ccfxyd genes were more dramatic and more irregular compared tothose in the group transferred from FW to SW, particularly for Ccnka

α1a and Ccfxyd2. Expression of these proteins was not assessed in thisstudy and thus remains to be elucidated.

5. Conclusions

This study revealed the potential modulatory mechanisms for NKAexpression from isoform compositions of NKA α-subunits and FXYDproteins in kidneys of the milkfish upon hypoosmotic (i.e. FW) chal-lenges. Four isoforms of milkfish NKA α-subunits with tissue-specificdistribution were identified: CcNKA α1a, α1c, α2, and α3. The highlyconserved amino acid sequences and similar characteristic features tohomologues from other animal species revealed their potential func-tion. Notably, in milkfish kidneys, CcNKA α1a and CcFXYD2 werepredominantly expressed. Increases in renal Ccnka α1a and Ccfxyd2genes were found after exposure of milkfish to FW. After transfer, mostrenal Ccnka and Ccfxyd genes (particularly Ccnka α1a and Ccfxyd2)clearly changed their expression, although most of these changes wererecovered subsequently. Our results suggest that the NKA activity ofmilkfish kidneys might be modulated via CcNKA α1a and CcFXYD2 for

Fig. 6. Time-course changes in mRNA expression of renal Ccnka α genes in the milkfish after transfer from fresh water (FW) to seawater (SW) (the study group) or toFW (the control group). The values are means± SE (N = 4 or 6 in the initial time-point or others, respectively). Different letters indicate significant differencesamong time- points in the study group (from FW to SW; Kruskal-Wallis test; P < .05). No difference was found among time-points in the control group (from FW toFW). The asterisks (*) indicate significant differences between the study and control groups at the same time-point post transfer (Mann-Whitney U test; P < .05).

W.-K. Yang et al. Aquaculture 504 (2019) 59–69

67

osmoregulation in response to salinity challenge. To our knowledge,this study provides the first evidence of the potential modulatory me-chanisms for NKA expression/activity from isoform compositions ofNKA α-subunits and FXYD proteins in kidneys of euryhaline teleosts.Further studies are necessary to illustrate more regulatory mechanismsof milkfish upon salinity challenge and the results are expected tobenefit and improve its aquaculture practice.

Acknowledgements

This study was financially supported in part by the Ministry ofScience and Technology (MOST), Taiwan, through the MOST ResearchProject (MOST-106-2313-B-005-038-MY3) and the MOST-CNRS(France) International Project for Scientific Cooperation (MOST-105-2911-I-005-501) grants to T.H.L. This work was also financially sup-ported in part by the iEGG and Animal Biotechnology Center from TheFeature Area Research Center Program within the framework of theHigher Education Sprout Project by the Ministry of Education (MOE),Taiwan (MOE-107-S-0023-F) to T.H.L. W.K.Y. was supported by post-doctoral fellowships from MOST (MOST-106-2811-B-005-004) andMOE (MOE-107-S-0023-F).

Conflict of interest

The authors have no competing interests to declare.

Appendix A. Supplementary data

Supplementary data to this article can be found online at https://doi.org/10.1016/j.aquaculture.2019.01.046.

References

Armesto, P., Campinho, M.A., Rodríguez-Rúa, A., Cousin, X., Power, D.M., Manchado, M.,Infante, C., 2014. Molecular characterization and transcriptional regulation of theNa+/K+ ATPase α subunit isoforms during development and salinity challenge in ateleost fish, the Senegalese sole (Solea senegalensis). Comp. Biochem. Physiol. B 175,23–38.

Arystarkhova, E., 2016. Beneficial renal and pancreatic phenotypes in a mouse deficientin FXYD2 regulatory subunit of Na,K-ATPase. Front. Physiol. 7, 88.

Bagarinao, T., 1994. Systematics, distribution, genetics and life history of milkfish,Chanos chanos. Environ. Biol. Fish 39, 23–41.

Blanco, G., 2005. Na,K-ATPase subunit heterogeneity as a mechanism for tissue-specificion regulation. Semin. Nephrol. 25, 292–303.

Blanco, G., Mercer, R.W., 1998. Isozymes of the Na-K-ATPase: heterogeneity in structure,diversity in function. Am. J. Physiol. Renal. Physiol. 275, F633–F650.

Blom, N., Sicheritz-Pontén, T., Gupta, R., Gammeltoft, S., Brunak, S., 2004. Prediction ofpost-translational glycosylation and phosphorylation of proteins from the amino acidsequence. Proteomics 4, 1633–1649.

Blondeau-Bidet, E., Bossus, M., Maugars, G., Farcy, E., Lignot, J.-H., Lorin-Nebel, C.,2016. Molecular characterization and expression of Na+/K+-ATPase α1 isoforms inthe European sea bass Dicentrarchus labrax osmoregulatory tissues following salinitytransfer. Fish Physiol. Biochem. 42, 1647–1664.

Bollinger, R.J., Madsen, S.S., Bossus, M.C., Tipsmark, C.K., 2016. Does Japanese medaka(Oryzias latipes) exhibit a gill Na+/K+-ATPase isoform switch during salinity change?J. Comp. Physiol. B. 186, 485–501.

Buchan, D.W.A., Minneci, F., Nugent, T.C.O., Bryson, K., Jones, D.T., 2013. Scalable webservices for the PSIPRED Protein Analysis Workbench. Nucleic Acids Res. 41,W349–W357.

Chang, C.H., Yang, W.K., Lin, C.H., Kang, C.K., Tang, C.H., Lee, T.H., 2016. FXYD11mediated modulation of Na+/K+-ATPase activity in gills of the brackish medaka(Oryzias dancena) when transferred to hypoosmotic or hyperosmotic environments.Comp. Biochem. Physiol. A 194, 19–26.

Chen, X.L., Wee, N.L.J.E., Hiong, K.C., Ong, J.L.Y., Chng, Y.R., Ching, B., Wong, W.P.,Chew, S.F., Ip, Y.K., 2013. Properties and expression of Na+/K+-ATPase α-subunitisoforms in the brain of the swamp eel, Monopterus albus, which has unusually highbrain ammonia tolerance. PLoS One 8, e84298.

Ching, B., Woo, J.M., Hiong, K.C., Boo, M.V., Choo, C.Y.L., Wong, W.P., Chew, S.F., Ip,Y.K., 2015. Na+/K+-ATPase α-subunit (nkaα) isoforms and their mRNA expressionlevels, overall Nkaα protein abundance, and kinetic properties of Nka in the skeletalmuscle and three electric organs of the electric eel, Electrophorus electricus. PLoS One10, e0118352.

Crambert, G., Geering, K., 2003. FXYD Proteins: new tissue-specific regulators of theubiquitous Na,K-ATPase. Sci. STKE 2003, re1.

Daly, S.E., Lane, L.K., Blostein, R., 1994. Functional consequences of amino-terminal

diversity of the catalytic subunit of the Na,K-ATPase. J. Biol. Chem. 269,23944–23948.

Dalziel, A.C., Bittman, J., Mandic, M., Ou, M., Schulte, P.M., 2014. Origins and functionaldiversification of salinity-responsive Na+, K+ ATPase α1 paralogs in salmonids. Mol.Ecol. 23, 3483–3503.

Evans, D.H., Claiborne, J.B., 2009. Osmotic and ionic regulation in fishes. In: Evans, D.H.(Ed.), Osmotic and Ionic Regulation. CRC Press, Boca Raton, FL, pp. 295–366.

Feng, S.H., Leu, J.H., Yang, C.H., Fang, M.J., Huang, C.J., Hwang, P.P., 2002. Gene ex-pression of Na+-K+-ATPase α1 and α3 subunits in gills of the teleost Oreochromismossambicus, adapted to different environmental salinities. Mar. Biotechnol. 4,379–391.

Geering, K., 2008. Functional roles of Na,K-ATPase subunits. Curr. Opin. Nephrol.Hypertens. 17, 526–532.

Hall, T.A., 1999. BioEdit: a user-friendly biological sequence alignment editor and ana-lysis program for Windows 95/98/NT. Nucleic Acids Symp. Ser. 41, 95–98.

Hu, Y.C., Chu, K.F., Yang, W.K., Lee, T.H., 2017. Na+, K+-ATPase β1 subunit associateswith α1 subunit modulating a “higher-NKA-in-hyposmotic media” response in gills ofeuryhaline milkfish, Chanos chanos. J. Comp. Physiol. B. 187, 995–1007.

Hu, P., Li, S., Zhong, Y., Mu, X., Gui, L., Zhang, J., 2014. Identification of fxyd genes fromthe spotted scat (Scatophagus argus): molecular cloning, tissue-specific expression,and response to acute hyposaline stress. Comp. Biochem. Physiol. B 174, 15–22.

Hwang, P.P., Lee, T.H., Lin, L.Y., 2011. Ion regulation in fish gills: recent progress in thecellular and molecular mechanisms. Am. J. Physiol. Regul. Integr. Comp. Physiol.301, R28–R47.

Ip, Y.K., Loong, A.M., Kuah, J.S., Sim, E.W.L., Chen, X.L., Wong, W.P., Lam, S.H., Delgado,I.L.S., Wilson, J.M., Chew, S.F., 2012. Roles of three branchial Na+-K+-ATPase α-subunit isoforms in freshwater adaptation, seawater acclimation, and active ammoniaexcretion in Anabas testudineus. Am. J. Physiol. Regul. Integr. Comp. Physiol. 303,R112–R125.

Kang, C.K., Chen, Y.C., Chang, C.H., Tsai, S.C., Lee, T.H., 2015. Seawater-acclimationabates cold effects on Na+, K+-ATPase activity in gills of the juvenile milkfish,Chanos chanos. Aquaculture 446, 67–73.

Kumar, S., Stecher, G., Li, M., Knyaz, C., Tamura, K., 2018. MEGA X: molecular evolu-tionary genetics analysis across computing platforms. Mol. Biol. Evol. 35, 1547–1549.

Liang, F., Li, L., Zhang, G., Yin, S., Wang, X., Li, P., Jia, Y., Wang, Y., Wang, L., Wang, X.,2017. Na+/K+-ATPase response to salinity change and its correlation with FXYD11expression in Anguilla marmorata. J. Comp. Physiol. B. 187, 973–984.

Liao, B.K., Chen, R.D., Hwang, P.P., 2009. Expression regulation of Na+-K+-ATPase α1-subunit subtypes in zebrafish gill ionocytes. Am. J. Physiol. Regul. Integr. Comp.Physiol. 296, R1897–R1906.

Lin, Y.M., Chen, C.N., Lee, T.H., 2003. The expression of gill Na, K-ATPase in milkfish,Chanos chanos, acclimated to seawater, brackish water and fresh water. Comp.Biochem. Physiol. A 135, 489–497.

Livak, K.J., Schmittgen, T.D., 2001. Analysis of relative gene expression data using real-time quantitative PCR and the 2−ΔΔCT method. Methods 25, 402–408.

Madsen, S.S., Engelund, M.B., Cutler, C.P., 2015. Water transport and functional dy-namics of aquaporins in osmoregulatory organs of fishes. Biol. Bull. 229, 70–92.

Marshall, W.S., Grosell, M., 2006. Ion transport, osmoregulation, and acid-base balance.In: Evans, D.H., Claiborne, J.B. (Eds.), The Physiology of Fishes. CRC Press, BocaRaton, pp. 177–230.

Martin, D.W., 2005. Structure-function relationships in the Na+,K+-pump. Semin.Nephrol. 25, 282–291.

Martinez, F.S., Tseng, M.C., Yeh, S.P., 2006. Milkfish (Chanos chanos) culture: situationsand trends. J. Fish. Soc. Taiwan 33, 229–244.

McCormick, S.D., Regish, A.M., Christensen, A.K., 2009. Distinct freshwater and seawaterisoforms of Na+/K+-ATPase in gill chloride cells of Atlantic salmon. J. Exp. Biol. 212,3994–4001.

Morth, J.P., Pedersen, B.P., Buch-Pedersen, M.J., Andersen, J.P., Vilsen, B., Palmgren,M.G., Nissen, P., 2011. A structural overview of the plasma membrane Na+,K+-ATPase and H+-ATPase ion pumps. Nat. Rev. Mol. Cell Biol. 12, 60–70.

Ogawa, H., Toyoshima, C., 2002. Homology modeling of the cation binding sites ofNa+K+-ATPase. Proc. Natl. Acad. Sci. U. S. A. 99, 15977–15982.

Richards, J.G., Semple, J.W., Bystriansky, J.S., Schulte, P.M., 2003. Na+/K+-ATPase α-isoform switching in gills of rainbow trout (Oncorhynchus mykiss) during salinitytransfer. J. Exp. Biol. 206, 4475–4486.

Saito, K., Nakamura, N., Ito, Y., Hoshijima, K., Esaki, M., Zhao, B., Hirose, S., 2010.Identification of zebrafish Fxyd11a protein that is highly expressed in ion-trans-porting epithelium of the gill and skin and its possible role in ion homeostasis. Front.Physiol. 1, 129.

Suhail, M., 2010. Na+, K+-ATPase: ubiquitous multifunctional transmembrane proteinand its relevance to various pathophysiological conditions. J. Clin. Med. Res. 2, 1–17.

Takei, Y., Hiroi, J., Takahashi, H., Sakamoto, T., 2014. Diverse mechanisms for body fluidregulation in teleost fishes. Am. J. Physiol. Regul. Integr. Comp. Physiol. 307,R778–R792.

Tang, C.H., Chiu, Y.H., Tsai, S.C., Lee, T.H., 2009. Relative changes in the abundance ofbranchial Na+/K+-ATPase α-isoform-like proteins in marine euryhaline milkfish(Chanos chanos) acclimated to environments of different salinities. J. Exp. Zool. A311, 521–529.

Tang, C.H., Lai, D.Y., Lee, T.H., 2012. Effects of salinity acclimation on Na+/K+-ATPaseresponses and FXYD11 expression in the gills and kidneys of the Japanese eel(Anguilla japonica). Comp. Biochem. Physiol. A 163, 302–310.

Tang, C.H., Wu, W.Y., Tsai, S.C., Yoshinaga, T., Lee, T.H., 2010. Elevated Na+/K+-ATPase responses and its potential role in triggering ion reabsorption in kidneys forhomeostasis of marine euryhaline milkfish (Chanos chanos) when acclimated to hy-potonic fresh water. J. Comp. Physiol. B. 180, 813–824.

Tipsmark, C.K., 2008. Identification of FXYD protein genes in a teleost: tissue-specific

W.-K. Yang et al. Aquaculture 504 (2019) 59–69

68

expression and response to salinity change. Am. J. Physiol. Regul. Integr. Comp.Physiol. 294, R1367–R1378.

Toyoshima, C., Kanai, R., Cornelius, F., 2011. First crystal structures of Na+,K+-ATPase:new light on the oldest ion pump. Structure 19, 1732–1738.

Vasilets, L.A., Schwarz, W., 1993. Structure-function relationships of cation binding in theNa+/K+-ATPase. Biochim. Biophys. Acta-Rev. Biomembr. 1154, 201–222.

Wang, P.J., Yang, W.K., Lin, C.H., Hwang, H.H., Lee, T.H., 2017. FXYD8, a novel regulatorof renal Na+/K+-ATPase in the euryhaline teleost, Tetraodon nigroviridis. Front.Physiol. 8, 576.

Whittamore, J.M., 2012. Osmoregulation and epithelial water transport: lessons from theintestine of marine teleost fish. J. Comp. Physiol. B. 182, 1–39.

Wong, M.K.S., Pipil, S., Ozaki, H., Suzuki, Y., Iwasaki, W., Takei, Y., 2016. Flexible se-lection of diversified Na+/K+-ATPase α-subunit isoforms for osmoregulation in tel-eosts. Zoological Lett. 2, 15.

Yang, W.K., Chung, C.H., Cheng, H.C., Tang, C.H., Lee, T.H., 2016a. Different expressionpatterns of renal Na+/K+-ATPase α-isoform-like proteins between tilapia and milk-fish following salinity challenges. Comp. Biochem. Physiol. B 202, 23–30.

Yang, W.K., Hsu, A.D., Kang, C.K., Lai, I.P., Liao, P.S., Lee, T.H., 2018. Intestinal FXYD12and sodium-potassium ATPase: a comparative study on two euryhaline medakas inresponse to salinity changes. PLoS One 13, e0201252.

Yang, W.K., Kang, C.K., Chang, C.H., Hsu, A.D., Lee, T.H., Hwang, P.P., 2013. Expressionprofiles of branchial FXYD proteins in the brackish medaka Oryzias dancena: a po-tential saltwater fish model for studies of osmoregulation. PLoS One 8, e55470.

Yang, W.K., Kang, C.K., Hsu, A.D., Lin, C.H., Lee, T.H., 2016b. Different modulatorymechanisms of renal FXYD12 for Na+-K+-ATPase between two closely related me-dakas upon salinity challenge. Int. J. Biol. Sci. 12, 730–745.

Yang, S.H., Tsai, J.D., Kang, C.K., Yang, W.K., Kung, H.N., Lee, T.H., 2017. The ultra-structural characterization of mitochondria-rich cells as a response to variations insalinity in two types of teleostean pseudobranch: milkfish (Chanos chanos) andMozambique tilapia (Oreochromis mossambicus). J. Morphol. 278, 390–402.

Zhu, H., Liu, Z., Gao, F., Lu, M., Liu, Y., Su, H., Ma, D., Ke, X., Wang, M., Cao, J., Yi, M.,2018. Characterization and expression of Na+/K+-ATPase in gills and kidneys of theTeleost fish Oreochromis mossambicus, Oreochromis urolepis hornorum and their hy-brids in response to salinity challenge. Comp. Biochem. Physiol. A 224, 1–10.

W.-K. Yang et al. Aquaculture 504 (2019) 59–69

69